A Detailed Analysis on Disaster Tweet Analysis Using Deep Learning Techniques: DTWEET

The widespread use of PDAs makes it easier for people to repeatedly declare a crisis. Twitter has developed into a significant correspondence platform in times of emergency. Despite the fact that states and crisis management organisations collaborate through their unique public frameworks for responding to disasters, the effectiveness of the disaster response and the recovery cycle is ultimately determined by the views of those affected during and after the event. When a crisis strikes, people publish frequent updates on online entertainment platforms. The vast amounts of unstructured data generated by online entertainment platforms like Twitter can be filtered using text mining and AI calculations to identify such data through watchwords and idioms that refer to disasters. This study focuses on using Normal Language Processing (NLP) and order models to distinguish between real and fake disaster tweets. The dataset was obtained from the Kaggle website and includes tweets that are related to actual disasters as well as other tweets that make references to fictitious disasters.